Journal of Structural Engineering Vol.55A (March 2009)
JSCE
Evaluation of seismic resistance for a multi-spans bridge in Vietnam by
investigation of earthquake activity and dynamic response analysis
Tran Viet Hung*, Osamu Kiyomiya** and Tongxiang An***
* Graduate student, Dept. of Civil and Environmental Eng., Waseda University, Ohkubo, Shinjuku-ku, Tokyo 169-8555
** Dr. of Eng., Professor, Dept. of Civil and Environmental Eng., Waseda University, Ohkubo, Shinjuku-ku, Tokyo 169-8555
*** Dr. of Eng., Research Associate of Advanced Res. Inst. for Sci. and Eng., Waseda University,
Ohkubo, Shinjuku-ku, Tokyo 169-8555
The aseismicity of a multi-spans highway bridge designed by Vietnam Specification is
evaluated by dynamic response analysis according to Japan Specification. The bridge
adopted in this study is located at Hanoi in Vietnam where is classified to a low moderate
seismic zone based on the investigation. As to Level 1 earthquake, there is no damage in the
bridge, i.e. the seismic performance of the bridge is also secured by Japan Specification. As
to Level 2 earthquake, though girder’s unseating is not calculated, the bridge pier is
predicted to have damage due to the large rotation angle at the plastic hinge and the lack of
the shear resistance of the pier.
Keywords: bridge in Vietnam, earthquake activity, dynamic response analysis, stopper
1. INTRODUCTION
Vietnam is located at moderated seismic activity area. However
seismic design for buildings and bridges become to be regarded
The arrangement of the social infrastructure is now rapidly
progressing with the social advance and economic development
as important design and so on in Vietnam. Seismic Design
Specifications for Bridges in Vietnam (a part of 22TCN 272-05)
in Vietnam. Recently, many bridges have been constructed and
was established based on AASHTO LRFD 1998 in 2001,
some bridges are still under constructing or planning. The bridge
takes an important role in road and railway networks. Of
officially applied in 2005. Recently many bridges are designed
by this Specification, including some bridges constructed under
particular note, the recent Northridge Earthquake of 1994, the
Hyogoken-Nanbu Earthquake of 1995, the Taiwan Chi-Chi
Japanese financial support (ODA).
As for an essential multi-spans bridge located moderate
Earthquake of 1999, the Iran Earthquake of 2001, the Chuetsu
Earthquake of 2004, and the Wengchuan Earthquake of 2008,
seismic zone, the seismic design can be conducted by
single-mode elastic method or uniform load elastic method
caused serious damage to many lifeline facilities, including
according to Vietnam Specification, however, the seismic design
bridges. Vietnam has not experienced the big earthquake
damages up to now and does not have the history of large scale
based on Japan Specification must be carried out by static
analysis for level 1 and level 2 earthquake, and dynamic analysis
earthquakes. The seismic designs for the bridges weren’t adopted
until former century 90's. Many seismometers are installed in
for level 2 earthquake, as to the seismic load used in the design is
modified by the zone factor. In addition, it is uncontradictable that
Vietnam and some middle scale earthquakes have been recorded
there is potentiality for large-scale earthquake to happen in
by the seismometers installed in Vietnam. According to the
analyses of the earthquake records obtained by the seismometers,
Vietnam. As a study, it is necessary to verify the seismic
performance of the bridges in Vietnam in case of a large-scale
-537-
earthquake. To evaluate the aseismicity of the bridges in Vietnam
and compare the difference between Vietnam Specifications for
central part of Vietnam. Table 3 is a catalog of some of strong
earthquakes in Hanoi region. The events from 1930 to the end of
Bridge Design and Japan Specifications for Highway Bridge, the
earthquake activity in Vietnam is firstly investigated and a
2006 are all the instrumental data of the seismological network of
the Northern Vietnam, China and ISC. Magnitude MS used in
multi-spans highway bridge designed by Vietnam Specification is
adopted as an example to compare the results for these two design
Table 3 is the surface wave magnitude; I0 is seismic intensity in
epicenter estimated following MSK scale that is similar to MM
codes.
scale. The error of magnitude estimation is about 0.2 - 0.3 and of
intensity I0 is about 0.3 - 0.5 2).
2. STATUS OF EARTHQUAKE AND SEISMIC DESIGN
FOR BRIDGE IN VIETNAM
2.2 Seismic design for bridge in Vietnam Specifications
Vietnam Specifications of bridge design (referred to as
2.1 Earthquake activities in Vietnam
Vietnam is situated between two major faults: Himalayan
22TCN-272-05) published in 2005 was established according to
AASHTO LRFD 1998. This specification also concerns with
Fault and Pacific Fault. These faults have strong effect on
seismic design, some objects are modified according to Vietnam
seismo-tectonic movement of this region. Moreover, Red River
fault running through the north portion of Vietnam has caused
conditions such as the classification seismic zone, map of
maximum seismic intensity zone, etc 3). However, the concept of
several earthquakes in this region. According to the earthquake
catalog compiled by the Vietnam Institute of Geophysics, 90% of
structure analysis and seismic analysis are originally taken from
AASHTO LRFD 1998.
the earthquakes took place in Vietnam is in Northwestern
Vietnam as shown in Fig. 1. Aside from this region, there was no
report of earthquakes of a magnitude larger than 5.5. Vietnam has
been classified as a low seismicity region.
Some of the significant earthquakes happened recently in this
region are presented in Table 1. The last two earthquakes caused
strong shakings in Hanoi. Most recently, on 19 February, 2001,
an earthquake of 5.3 Richter scale was occurred in Dienbien
(close to the border with Laos PDR) at a depth of 12 kilometers.
This earthquake was caused by the Laichau - Dienbien rupture,
parallel with the Hong (Red) River rupture. Although the
earthquakes happened up to now are moderate ones it is predicted
that the Hong (Red) River rupture has potential of causing
earthquakes of 7 - 8 Richter scale. In case those large scale
earthquakes happen, the buildings and structures in this area of
thousands square kilometers may be destructed 1).
In Vietnam, during the 20 Century 2 earthquakes with
intensity I0 = 8 - 9 (MSK-64) and magnitude M = 6.5 - 7 Richter
degrees, 15 earthquakes with I0 = 7 and M = 5 - 5.9, and more
Fig. 1 Map of active faults and earthquake epicenters distribution
in Vietnam from 1067 to 2002 (Nguyen, 2007).
than 100 earthquakes with I0 = 6 - 7 and M = 4.5 - 4.9 were
observed according to “Research and Forecasting Earthquakes
and Foundation Fluctuations in Vietnam” (Prof. Nguyen Dinh
Table 1 Catalog of strong earthquakes in Vietnam
Xuyen and his collaborators, Vietnam Institute of Geophysics).
No
Year
As shown in Table 2, earthquakes of 5.5 - 6.8 on the Richter’s
scale would happen at 30 areas in Vietnam. In Hanoi capital, the
1
1961
tectonic fault exists along the Hong (Red) River, Chay River and
Lo River. The seismographic survey shows that earthquakes of
2
3
1983
1987
magnitude MS = 5.0 have occurred in North Vietnam, in the
north border of Hanoi depression and in the coastal zone of the
4
5
1988
2001
Campha
Dienbien
-538-
Depth
Magnitude
Intensity
(km)
MS
I0
Bacgiang
28
5.5 - 5.9
7
Tuangiao
Nhanam
13
10
5.3
4.6
7
6
10
12
4.9
5.3
7
7
Location
Table 2 List of areas in risk for strong earthquakes in Vietnam (Nguyen Dinh Xuyen et al., 2005)
No.
Maximum earthquake
Area
No.
(Richter)
Maximum earthquake
Area
(Richter)
1
2
Son La
Dong Trieu
6.8
6.0
16
17
Song Ma - Fumaytun
Red River - Chay River
6.5
6.0
3
Song Ca - Khe Bo
6.0
18
Rao Nay
5.5
4
5
Cao Bang-Tien Yen
Cam Pha
5.5
5.5
19
20
Hanoi’s Northeastern sunken area
Lo River
5.5
5.5
6
Phong Tho - Than Uyen
Muong La - Cho Bo
5.5
21
Da River
5.5
7
8
Muong Nhe
Hieu River
5.5
5.5
22
23
Ma River’s lower section
Khe Giua - Vinh Linh
5.5
5.5
9
Tra Bong
5.5
24
Hue
5.5
10
11
Da Nang
Poco River
5.5
5.5
25
26
Tam Ky - Phuoc Son
Ba River
5.5
5.5
12
13
Ba To - Cung Son
Tuy Hoa - Cu Chi
5.5
5.5
27
28
109.5 meridian
Thuan Hai - Minh Hai
5.5
5.5
14
Vung Tau - Ton Le Sap
5.5
29
Hau River
5.5
15
Phu Quy 1
5.5
30
Phu Quy 2
5.5
Table 3 Catalog of some strong earthquakes near site construction (Hanoi, Vietnam) to the end of 2006 2)
No
Year
Depth
(km)
Magnitude
MS
Intensity
I0
No
Year
Depth
(km)
Magnitude
MS
Intensity
I0
1
1934
17.1
5.0
6
6
1961
28 ± 7
5.6 ± 0.4
7 ± 0.3
2
3
1939
1942
12.8
7±1
4.7
4.4 ± 0.2
6 ± 0.3
6 ± 0.3
7
8
1967
1974
20
17 ± 13
5.0 ± 0.4
4.1 ± 0.7
6 ± 0.3
5 ± 0.3
4
1945
11
4.7 ± 0.5
6 ± 0.3
9
1989
5-6
4.9
7
5
1958
20 ± 5
5.3 ± 0.3
6 ± 0.3
10
2005
-
4.7
-
Earthquake loads are given by the product of the mass and the
Table 4 Seismic zones in Vietnam
Acceleration Coefficient
Seismic zone
MSK - 64 class
A ≤ 0.09
0.09 < A ≤ 0.19
0.19 < A < 0.29
1
2
Class ≤ 6.5
6.5 < Class ≤ 7.5
3
7.5 < Class ≤ 8
elastic seismic response coefficient Csm for the mth vibration
mode. The coefficient is given as following equations.
Csm =
S
(Tm ≤ 4.0 s)
(1a)
3 AS
Tm4 / 3
(Tm > 4.0 s)
(1b)
Csm =
Table 5 Site coefficients (22TCN 272-05)
Site coefficient
1.2 AS
≤ 2.5 A
Tm2 / 3
Soil profile type
where Tm is the period of the mth vibration mode (s); A is the
I
II
III
IV
1.0
1.2
1.5
2.0
acceleration coefficient and it is determined in accordance with
the map of seismic zones and maximum seismic intensity zone
The seismic zone in Vietnam is classified into three zones as
presented in Table 4. The map of acceleration coefficient in
Vietnam is shown in Fig. 2 4). It is given by Vietnam Institute of
Geophysics with return period about 500 years.
of Vietnam (reference to Fig. 2) and, it is provided as contour for
return period of 500 years. Maximum probable earthquake with a
return period of around 2.500 years has to be considered with the
critical bridges. S is the site coefficient as shown in Table 5. The
four descriptive soil types are defined as follows 3):
-539-
+ Type I (S = 1.0): Rock of any characteristic or any stable
deposit of sands, gravels, or stiff clays less than 60 m deep and
overlying rock.
+ Type II (S = 1.2): Deep cohesion-less soil including any
stable deposit of sands, gravels, or stiff clays greater than 60 m
deep and overlying rock.
+ Type III (S = 1.5): Soft to medium stiff clay, sand, or other
cohesion - less soil generally greater than 9 m deep.
+ Type IV (S = 2.0): Soft clays or silts greater than 12 m in
depth.
In the seismic design in Vietnam, except for some especial
bridges, the ordinary bridges are design by static method. The
design sectional forces and displacements are calculated by acting
the earthquake load as static force and using liner static analysis on a
beam model. As for the important especial bridges such as Nhattan
Bridge (Cable stayed bridge), etc, the seismic design is conducted
based on such as Japan Specifications for Highway Bridge (referred
to as JRA-2002) to secure more safety of the bridges.
Fig. 2 Seismic zones and maximum seismic intensity zone of
Vietnam with return period about 500 years 4)
3. CONSIDERING SOME CHARACTERISTICS IN
SEISMIC DESIGN BETWEEN 22TCN-272-05 AND
JRA-2002
In seismic design, it is thought that the peak value of design
horizontal seismic coefficient is very important. Focusing on the
determination method of this coefficient for moderate level
earthquake, 22TCN-272-05 and JRA-2002 are compared. As to
22TCN-272-05, from equation (1a) and (1b), elastic seismic
response coefficient can be presented as follows:
Csm = f (A, S, Tm)
(2)
The design horizontal seismic coefficient for level 1
5)
earthquake according to JRA-2002 is as follows :
kh = Cz kh0
(3)
where kh0 = f (TG, T) as presented in Table 7.
It is found from (2) and (3) that there is similarity in the
seismic coefficient determined by the seismic zone, surface
ground condition and the vibration characteristics of the bridge
for both 22TCN-272-05 and JRA-2002:
- Seismic zone: modification factor Cz and acceleration
coefficient A.
- Ground type: TG (Type I, II, III) and site coefficient S = I, II,
III, IV.
- Natural period: T and Tm
- Moderate earthquake Level 1 in Japan is similar to moderate
earthquake in Vietnam.
The difference between these two codes is the values of the
parameters. As to the PGA, 0.16 - 0.24 is adopted in JRA-2002,
but 0.00 - 0.29 is used in 22TCN-272-05. As an example, some
of the PGA values provided by Vietnam Institute of Geophysics
for some construction sites are presented in Table 6. Table 8
shows some values of input data characteristics in seismic design
codes at both JRA-2002 and 22TCN-272-05.
Base on above observation, it is clear that some values of
seismic design are similar but earthquake characteristics and
seismic modification factors are different in these two codes. It is
significant to evaluate the seismic resistance of bridge with two
codes (JRA-2002 and 22TCN-272-05).
4. DESCRIPTION OF SEISMIC ANALYSIS PROCEDURES
4.1 Bridge structure
Fig. 3 shows the profile of a multi-spans continuous existing
bridge in Vietnam. The superstructure is a hollow slab beam
structure with 8 spans. The total length of the bridge is 250 m.
The surface layer consists of medium sand, fine sand and gravelly
sand as shown in Fig. 4. The thickness of the surface layer is
about 40 m. The cross section of girder is shown in Fig. 5. The
bridge consists of 3 rigid frame piers (P2, P3, and P4) and 4 bent
piers. Rubber bearing supports (P0, P1, P5 and P6) are installed.
The basic components of rubber bearing are elastomer and steel
plates as shown in Fig. 6. The pier columns and the abutments are
-540-
Table 6 Acceleration coefficient used to seismic design of some bridges in Vietnam
No
Name of Bridge
Typical Bridge
Span Layout
MSK-64
Acceleration
Class
coefficient (g)
1
Tan De
Cantilever Bridge
75+3@120+70
8
0.10
2
Phu Dong
Cantilever Bridge
65+7@100+65
7(8)
0.17
3
4
Bai Chay
Kien
Cantilever Bridge
Cable stayed Bridge
40+81+129+435+129+86
85+200+85
6
7
0.17
0.06
5
Can Tho
Cable stayed Bridge
2@40+150+550+150+2@40
6
0.10
6
7
Thanh Tri
Đa Bac
Cantilever Bridge
Cantilever Bridge
80+4@130+80
65+100+65
8
7
0.17
0.07
8
9
Non Nuoc
My Thuan
Cantilever Bridge
Cable stayed Bridge
42+52+85+52+42
150+350+150
7(8)
6
0.10
0.10
10
11
Ben Luc
Nhat Tan
Cantilever Bridge
Cable stayed Bridge
50+90+120+90+50
150+4@300+150
7
7(8)
0.10
0.12
12
Phu Long
Cantilever Bridge
75+120+75
7
0.08
13
Đong Tru
CFST arch bridge
80+120+80
8
0.17
Table 7 Characteristics of JRA-2002 with Level 1
kh0, value in term of natural period, T (s)
Ground type
Characteristic value of ground, TG (s)
Type I
kh0 = 0.2
TG < 0.2
Type II
kh0 = 0.25
0.2 ≤ TG < 0.6
Type III
kh0 = 0.3
0.6 ≤ TG
Table 8 Seismic design conditions according to Japan and Vietnam specifications
Japan Specifications for highway bridge
(JRA-2002)
Items
Vietnam Specifications for bridge design
(22TCN-272-05)
Seismic zones
Modification factor: Cz = 0.7 (region C)
Acceleration coefficient: A = 0.1 g (region 2)
Ground characteristics
Dynamic characteristic value of the surface ground:
TG = 0.82 s > 0.6 s Type III ground (soft ground
Site coefficient: Type III, S = 1.5 (Soft to medium
stiff clay, sand, or other cohesion-less soil generally
or alluvial ground).
greater than 9m deep)
Damping ratio
Pier (NLE): 2%, pier (LE): 5%, girder (LE): 3%,
bearing spring: 4%, footing: 10%
Seismic coefficient
Design horizontal seismic coefficient: kh = cz kh0
Elastic seismic response coefficient: Csm
Fig. 3 Profile of a highway bridge (unit is mm)
-541-
3.50
2.90
Lean clay
Lean clay
4.00
Lean clay
Fine sand
Medium sand
3.00 2.90
3.40
Medium sand
5.30
Silt clay
5.10
5.90
Type soil
Soft ground
10.40
Stratum
Fine sand
N - Value
1020 30 40 50
Pebble
Fig. 4 Borehole log
-fabricated by reinforced concrete and bored cast-in-place piles
Table 9 Characteristic value of the surface ground
are driven under the footing. The profile of the pier and the
foundation is shown in Fig. 7.
Layer
4.2 Analytical modeling
The bridge used in this study is designed by static analysis
according to the Vietnam design code. The stopper and rubber
bearing are installed to enhance the seismic performance.
According to the seismic design specified in Japan Specifications
for Highway Bridge, a nonlinear analysis model of the entire
bridge system as shown in Fig. 8 is made to estimate seismic
performance of the bridge, especially bearing capacity of the
members and unseating of the girders. Fig.9 shows detail model
of the pier with the foundation, the stopper and the rubber bearing.
hi (m)
SPT (N)
γt
3
(kN/m )
Vsi (m/s)
Ti =
Hi/Vsi (s)
Silt clay
5.9
3
17.4
138.67
0.04
Sand
10.4
10
16.0
218.28
0.05
Lean clay
3.4
3
18.3
135.72
0.03
Fine sand
5.3
21
16.8
273.69
0.02
Lean clay
5.1
3
16.5
144.22
0.04
Lean clay
2.9
12
19.7
228.94
0.01
Sand
4.0
20
16.8
271.44
0.01
Fine sand
2.9
24
16.8
288.45
0.01
Pebble
3.0
50
21.0
368.40
0.01
The girders are replaced to linear beams.
The pile foundation is replaced to the horizontal spring Kx,
T = 0.82 (s) > 0.60 (s)
Type III ground according to JRA-2002
vertical spring Ky and rotating spring Kθ. The spring values are
calculated by Forum 8 software. The dominate period of the
surface ground is TG = 0.82 s > 0.60 s i.e. ground type in Table 9
is type III ground. The modified factor Cz is selected as have 0.7
(region C). The concrete block as a stopper is installed at the top
of the pier as shown in Fig. 10. The stopper is replaced to a spring
element considering the spacing. The rubber bearing is replaced
to a bi-linear spring in horizontal direction. The pier has circler
cross section as shown in Fig. 11.
The compressive strength of the concrete is 30MPa, the
Fig. 5 Cross section of the girder (unit is mm)
diameter of the spiral reinforcement is 16mm and spacing of the
spiral is 300mm. The column of the pier is replaced to nonlinear
beam elements. The nonlinear behavior of the columns is
presented by the Takeda model with the potential plastic hinge
zone located at bottom of the column. The Takeda hysteresis
property is adopted for bending deformation of the pier.
Relationship of M - φ and M - θ for column piers are shown in
Fig. 12. The stress vs. strain relation of reinforcing bars is
idealized by bi-linear model.
Fig.6 Dimensions of rubber bearing (unit is mm)
In the pier column, a plastic hinge modeled by nonlinear
rotating spring is arranged. The length of the plastic hinge is
calculated by following equations and the results are presented in
Table 10.
Lp = 0.2h - 0.1D
in which: 0.1D
≦L ≦ 0.5D
p
(4a)
(4b)
where Lp is plastic hinge length; h is height of the column pier; D
is section depth (D is diameter of a circular section). The
Rayleigh damping coefficients are calculated from the vibration
mode of the structure.
-542-
Fig.7 Profile of pier and foundation (unit is mm)
A1
P0
P1
P3
P2
P5
P4
P6
A2
Fig. 8 Modeling of the entire bridge
Girder (linear beam element)
longitudinal
transverse
Bearing spring
Stopper
Acting force
K=10 kN/m
foamed styroll
Disp.
Fig. 10 Concrete stopper (unit is mm)
Disp.
-1
K=10 kN/m
2500
2300
2 D16
1
Colunm pier
Nonlinear beam element
Plastic hinge spring
Nonlinear rotating
spring element
Footing
Beam element
100
D25
k
Long.=0.12m
Hozi.=0.05m
Rubber buffer
100
k1
8
Spacing
foamed styroll
P
2 D16
Foundation ground spring
3 D16
3 D16
D16
D16
3
3
Fig. 9 Modeling of the bridge pier
Fig. 11 Cross section of pier column (unit is mm)
25000
25000
20000
)
.mN
k( 15000
tn
e 10000
mo
M
20000
)
.mN
(k 15000
tn
e 10000
mo
M
P1, P5
P0, P6
P2, P4
P3
5000
5000
φ (1/m)
0
0
01
0.0
02
0.0
03
0.0
04
0.0
05
0.0
06
0.0
P1, P5
P0, P6
P2, P4
P3
θ (rad)
0
0
07
0.0
a) Moment vs. curvature relation
5
00
0.0
5
01
01
0.0
0.0
5
02
02
0.0
0.0
5
03
03
0.0
0.0
5
04
04
0.0
0.0
b) Moment vs. rotation relation at the plastic hinge
Fig. 12 M - φ and M - θ relationships of column piers
Natural dominate frequencies of the structure are 2.337 Hz and
Newmark-β method and integration time interval is 0.01 s. The
7.305 Hz. The values of the Rayleigh damping coefficients are α
= 0.45829 and β = 0.00225, respectively. The commercial finite
damping ratios for members are presented in Table 11.
element analysis program (TDAP III software) is used for the
analysis. The numerical integration is performed using the
4.3 Input motions for dynamic response analysis
Two ground acceleration records in Japan are adopted as input
-543-
Table 10 Plastic hinge length of pier, Lp (m)
(*)
Pier
P0, P6
P1, P5
P2, P4
h (m)
3.9
4.8
5.7
Lp m)
0.5
0.7
(*)
180
0.4
(*)
P3
0.5
140
2
0.5
The piers P2, P3, P4 have 2 locations of the plastic hinge length
at bottom and top of pier
Table 11 Damping ratio of the members
)
s/ 120
m
(c 100
no
tia
re 80
le
cc 60
A
Level 1
(Cz=0.7),
(985 , 98.7)
40
Member
Damping ratio
Bridge column - pier (nonlinear member)
Bridge column - pier, footing (linear member)
2%
5%
Girder (linear member)
3%
Bearing spring
Foundation spring
4%
10%
20
0
0
2000
4000
6000
8000
10000
Return Period, Rd (Years)
Fig. 14 PGA with return periods at the near construction site in
Hanoi, Vietnam
Table 12 Intensities and PGA at site construction 2)
Return Periods
(Cz=1.0)
(6575 , 141)
5.9
0.4
Level 1
160
-data at the ground level. Namely, Tsugaru Ohhashi (1983) and
Site construction
(years)
A (cm/s )
I0 (MSK)
Kushirogawa (1994) as shown in Fig. 13 are adopted here. The
peak ground accelerations of them are 1.41 m/s2 and 4.38 m/s2,
1
2
200
500
67.38
85.56
6
6-7
respectively. These peak ground accelerations are corresponded
to Level 1 ground motion and Level 2 ground motion according
3
1000
100.26
7
4
5
2000
5000
116.53
138.24
7
8
to Seismic design for bridge in Japan Specifications (JRA-2002),
respectively. These ground acceleration records are adopted
6
10000
155.84
8
No
2.0
) 1.5
2
s/
1.0
(m
no 0.5
tia 0.0
re
le
0
cc -0.5
A-1.0
2
because the soil condition of the construction site is classified into
Group III in soil condition.
Table 12 shows the intensity (MSK-64 scale) and PGA with
different return periods at the construction site of this bridge. This
table is calculated from earthquake events data as presented in
Max acc. is 1.41m/s2
Time (s)
10
20
30
40
Table 2 and Table 3.
Fig. 14 shows relationship between the PGA (cm/s2) and
return periods at the near construction site of the bridge. For Level
50
1 earthquake motion, modified factor Cz = 0.7 is adopted because
this area is located at moderate earthquake activity. Only Level 1
-1.5
earthquake is presumed to occur and Level 2 is not presumed to
occur in this region. However, return period of Level 1 is much
a) Horizontal acc. in the Tsugaru Ohhashi, 1983
(Mg = 7.7, the maximum acc. is 1.41 m/s2)
6.0
Max acc. is 4.38m/s2
4.0
)
2
s/ 2.0
(m
.c 0.0
10
20
c -2.0 0
A
-4.0
-6.0
more at site construction yet and return period is 985 year for Cz =
0.7 and 6575 year for Cz = 1.0, respectively. While return period
of level 1 (JRA-2002) is about 50 to 100 years i.e., there is large
Time (s)
30
40
50
b) Horizontal acc. in the Kushirogawa, 1994
(Mg = 8.2, the maximum acc. is 4.38 m/s2)
Fig. 13 The ground acceleration records
60
difference as to return periods of Level 1 earthquake motion
between Vietnam and Japan. For Level 2 earthquake, its return
period is extremely long.
From view point of earthquake activities in Hanoi, Level 2
earthquake motion may not be considered in the design work.
However the damage of this bridge for Level 2 earthquake is
studied here to confirm the damage level. It is maybe important to
evaluate for inland earthquake about Mg = 6.5 - 7.0 as Level 2
earthquake motion in the future.
-544-
0.015
Pier P0
Pier P1
Pier P2
Pier P3
0.010
) 0.005
m
(t
en
m
ec 0.000
lap
si
D-0.005
Stopper
3
Non-stopper
2
2
Time (s)
0
10
20
30
40
50
60
)
s/ 1
(m
nio
ta 0
rla
ec
cA-1
-0.010
-2
-0.015
-3
Time (s)
0
10
a) Displacement vs. time relation at the top of the piers
) 0.025
(m
tn
e 0.02
m
ec
lap0.015
si
Dl 0.01
at
no0.005
izr
o 0
H
Non stopper
20
30
40
50
60
a) Acceleration of the girder end (A1)
Stopper
0.025
Stopper
Non-stopper
0.020
0.015
) 0.010
(m
tn 0.005
e
m
ec 0.000
alp-0.005
si
D-0.010
P0
P1
P2
P3
P4
P5
0
10
20
30
40
50
-0.015
P6
-0.020
b) Considering displacement at the top of the piers for case of
stopper and non-stopper
-0.025
b) Displacement of the girder end (A1)
Fig.15 Displacement at the top of the piers for Level 1
Fig. 16 Response of the bridge for Level 1
8000
8000
6000
Time (s)
Pier P1
6000
) 4000
m
.N 2000
k(
tn 0
e
m
o-2000
M-4000
) 4000
m. 2000
Nk
(
tn 0
e
mo -2000
M
-6000
-6000
Pier P2
-4000
-8000
-0.0002 -0.0001
0
0.0001
Rotation, θ (rad)
-8000
-0.0001 -0.00005
0.0002
0
0.00005
0
0.0002
Rotation, θ (rad)
0.0001
a) Case of stopper
15000
15000
10000
Pier P1
10000
)
m. 5000
Nk
(
tn
0
e
mo -5000
M
)
m. 5000
Nk
(
tn
0
e
mo
M -5000
-10000
-10000
-15000
-0.0004 -0.0002
0
0.0002
Rotation, θ (rad)
-15000
-0.0004
0.0004
Pier P2
-0.0002
Rotation, θ (rad)
b) Case of non-stopper
Fig. 17 Hysteretic response at the plastic hinge for Level 1
-545-
0.0004
60
5. RESPONSE OF BRIDGE STRUCTURE IN A LOW-
residual displacement limit of the pier δ/h ≤ 1/100 is not satisfied.
-MODERATE SEISMIC ZONE (LEVEL 1)
Where δ is displacement at top of the pier, h is height of the pier.
Following to 22TCN-272-05, longitudinal stoppers shall be
5
designed for a horizontal force calculated by the acceleration
coefficient and the self weight of the girder. Sufficient slack shall
be allowed in the stopper so that the restrainer does not start to act
until the design displacement is exceeded. The small
displacements of girder at pier P2, P3 and P4 are calculated as
shown in Fig. 15. Fig. 16 shows the response of the bridge for
PGA of the input wave is 1.41 m/s2. Results are compared as to
Non stopper
4
3
2
) 2
/s
(m 1
onti 0
ar
lee -1
cc
A-2
0
10
show that no damage is calculated under a moderate ground
excitation In this case, the stopper plays a not significant role for
50
60
Time (s)
-5
a) Case of non-stopper
30
non-stopper, respectively. Fig. 17 shows the hysteretic response
at the plastic hinge of the pier P1 that rubber bearing is installed at
stopper and non-stopper, respectively. The displacement at the top of
the pier is also small regardless of existing of the stopper. The results
40
-4
the girder are 1.4 cm and 2.2 cm for case of stopper and
4.149×10-5 rad and 3.134×10-4 rad, 2.062×10-4 rad for case of
30
-3
existing of the stopper. The maximum accelerations of the girder
are 1.62 m/s2 and 2.47 m/s2 while the maximum displacements of
top and pier P2 that is rigid jointed between the pier and the girder.
The maximum rotation angles of P1 and P2 are 1.072×10-4 rad,
20
Stopper
20
2
) 10
s/
m
( 0
no
it
ar -10
lee
cc
A-20
0
10
20
30
40
50
60
-30
Time (s)
-40
the damage of the bridge.
b) Case of stopper
Fig.18 Acceleration of the girder end (at A1)
6. RESPONSE OF THE BRIDGE WITH STRONG
GROUND EXCITATION (LEVEL 2)
0.30
Fig. 18, Fig. 19, Fig. 20 and Fig. 21 show the response of the
girder when the peak ground acceleration is 4.38 m/s2. The
maximum accelerations of the girder are 38.6 m/s2 and 4.52 m/s2
while the maximum displacements of the girder are 10.13cm and
23.91 cm for case of stopper and non-stopper, respectively. The
acceleration of the girder in case of the stopper is generally small
due to the spacing but very large acceleration is generated when
the girder movement stops suddenly due to the collision.
Longitudinal displacement of the girder can be restrained by the
Non stopper
Stopper
0.25
)
(m
tn0.20
e
m
ec
lap0.15
isd
la 0.10
tn
oz 0.05
ir
oH
0.00
Time (s)
0
10
20
30
40
50
60
-0.05
Fig. 19 Response of the girder for Level 2
stopper. When the stopper is not installed, fairly large
displacement of about 24 cm is calculated.
Fig. 22 shows the hysteretic response at the plastic hinge of the
pier P1 (rubber bearing) and pier P2 (rigid pier and girder). The
maximum rotation angles of P1, P2 are 3.15×10-2 rad, 3.67×10-2
rad and 4.06×10-2 rad, 8.34×10-2 rad for case of stopper and
non-stopper, respectively. The displacements at the top of the pier
are large with both cases. The maximum displacements at the top
of pier are 10.13 cm and 23.91 cm for case of stopper and
non-stopper, respectively. Almost piers have damage and the
-546-
) 0.3
m
( 0.25
tn
e
m
ec 0.2
lap
is 0.15
D
la 0.1
tn
oz 0.05
ir
o
H 0
Non stopper
P0
P1
P2
P3
P4
P5
Stopper
P6
Fig. 20 Displacement at the top of the piers for Level 2
70
Pier P0
0.12
Pier P1
Pier P2
Pier P3
0.10
-0.02
Pier P1
Pier P2
Pier P3
0.25
0.08
) 0.06
(m
nte 0.04
m
cea 0.02
lp
si 0.00
D
Pier P0
0.30
P1
P2
P3
P0
Time (s)
0
10
20
30
40
50
60
70
P1
) 0.20
(m
nte 0.15
m
cea 0.10
lp
si
D 0.05
P3
P0
Time (s)
0.00
-0.04
-0.06
0
10
20
30
40
50
60
-0.05
a) Case of stopper
b) Case of non-stopper
Fig. 21 Displacement at top of pier for Level 2
Resistance
30000
Response
Mc=6520.4kN.m
-0.03
Mc,θc
My,θy
-40000
Rotation, θ (rad)
-4
20000
-3
10000
θu=2.39x10 rad
Mu,θu
My,θy
Mc,θc
0
-0.02
-0.01
Mu,θu
-30000
Pier
P1
)
.mN Mu=22276.6kN.m
k(
tn -0.04
-0.03
e
mo
M
0.01
-20000
30000
-5
θc=4.92x10 rad
My=20895.1kN.m θy=5.34x10 rad
Mc,θc
0
-0.01
0
-10000
-0.02
40000
Response
Mc=6569.2kN.m
My,θy
My=20756.4kN.m θy=9.29x10-4rad 10000
Mu=22224.0kN.m θu =4.18x10-3rad
)
.mN
k(
tn-0.04
em
o
M
Resistance
Mu,θu
20000
θc =8.54x10-5 rad
-10000
0
0.01
Mc,θc
-20000
My,θy
-30000
Mu,θu
-40000
Pier
P2
-50000
Rotation, θ (rad)
a) Case of stopper
Resistance
30000
Response
20000
-5
Mc=6520.4kN.m θc=8.54x10 rad
My=20756.4kN.m θy=9.29x10-4rad
Mu=22224.0kN.m θu=4.18x10-3rad
)
m.
Nk
(
tn -0.05
e
mo
M
-0.04
-0.03
10000
Mc
0
-0.01
0
-10000
-0.02
-20000
-30000
Pier
P1
Rotation, θ (rad)
Mc=6569.2kN.m θc=4.92x10-5rad
My=20895.1kN.m θy=5.34x10-4rad
Mu=22276.6kN.m θu=2.39x10-3rad
My Mu
)
.mN -0.1
k(
tn
e
mo
M
0.01
Mc
-0.08
-0.06
Resistance
-0.04
Response
30000
10000
-0.02
-10000 0
-30000
Mu
My
Mc
Mc
My
Mu
-70000
-90000
Pier
P2
Rotation, θ (rad)
b) Case of non-stopper
Fig. 22 Hysteretic response at the plastic hinge for Level 2
Table 13 Summary of the results at P1 pier
JRA-2002 (Response)
Level 1
Level 2
22TCN-272-05
(Resistance)
Cz = 0.7
Cz = 1.0
Cz = 0.7
Bending moment (kN·m)
Shear resistance (kN)
6914
1588
9423
2163
21250
4841
18538
3276
Displacement of pier (m)
0.007
0.015
0.094
0.048
-547-
0.02
-50000
My
Mu
-40000
Items
P2
70
×
uG = εG L
7. EARTHQUAKE EFFECTS TO EXISTING BRIDGE
7.1 Shear and moment resistance of the pier
For shear resistance capacity of the pier, both JRA-2002 and
22TCN-272-05 are provided by shear strength of a reinforced
concrete pier which consists of the concrete section and the spiral
section. In 22TCN-272-05, the shear resistance capacity (Vr) is
calculated by the equation of Vr ≤ φVn, where φ = 0.85 is
resistance factor. The nominal shear resistance Vn shall be
determined as the lesser of bellow two equations.
Vn = Vc + Vs
(5a)
Vn = 0.25 f’c bv dv
(5b)
(6c)
where uR is the relative displacement developed between a
substructure and a superstructure (m), uG is relative displacement
of the ground along the longitudinal direction of a bridge (m), SEM
is a minimum seat length (m), εG is ground strain induced during
an earthquake along the longitudinal direction of the bridge (equal
to 0.005 for groups III ground conditions), L is a distance
contributing to the relative displacement of the ground (m), and l
is a span length (m). The seat length (JRA-2002) designed at the
abutment A1 is:
×
×
SE = 0.24 + 0.005 125 = 0.865 m ≥ SEM = 0.7 + 0.005 30
= 0.85 m.
in which:
To account for the possibility of plastic hinging and the
f c' b v d v
Vc = 0.083 β
Vs =
A v f y d v (cot gθ + cot gα ) sin α
s
(5c)
associated large displacement, 22TCN-272-05 requires that the
beam seat length (N) shall be taken empirical displacement
formula as follows:.
N = (200+0.0017L+0.0067H)(1+0.000125S2)
(5d)
(7)
where N is minimum support length measured normal to the
where bv is effective web width (mm), dv is effective shear depth
centerline of bearing (mm); L is length of the bridge deck to the
(mm), s is spacing of stirrups (mm), β is factor indicating
ability of diagonally cracked concrete to transmit tension, θ is
adjacent expansion joint, or to the end of the bridge deck; for
hinges within a span, L shall be the sum of the distances to either
angle of inclination of diagonal compressive stresses (deg), α is
angle of inclination of transverse reinforcement to longitudinal
side of the hinge; for single-span bridges, L equals the length of
the bridge deck (mm); H is average height of columns supporting
axis (deg), Av is area of shear reinforcement within a distances
(mm2). The flexural resistance capacity (Mr) is calculated by the
the bridge deck to the next expansion joint (mm) for columns
and/or piers, column, or pier height (mm) for hinges within a span,
equation of Mr ≤ φ·Mn, where φ = 0.85 is resistance factor.
average height of the adjacent two columns or piers (mm) for
Take the typical pier P1 as an example to observe, besides for
extreme capacity of pier, the case of modified factor Cz = 1.0 is
also considered. The results shows the pier has no damage to low
moderate earthquake level 1 in both case of modified factor Cz =
single-span bridges (mm); S is skew of support measured from
line normal to span (deg).
The empirical seat length (22TCN-272-05) designed at the
abutment A1 is N = 0.88 m.
0.7 and Cz = 1.0 as shown in Table 13. However the pier has
severe damage by strong earthquake level 2.
If bearing capacity of the pier is satisfied for Japanese code,
longitudinal reinforcement of D25 is replaced to D32 for bending
moment and spiral D16 with 300 mm spacing is replaced to D19
with 150 mm spacing.
7.2 Evaluation of seating length
The seat length required to prevent unseating is considered.
The seat length SE (m) is determined from the Japanese Design
Fig. 23 Girder end support (mm)
Specifications for Highway Bridges as follows:
SE =uR + uG ≥ SEM
×
SEM = 0.7 + 0.005 l
(6a)
The relative displacement between the substructure and the
(6b)
superstructure is defined by dynamic response analysis is 21.5 cm
for level 2. This value is within both design values. Unseating of
-548-
the girder will not be happened. Fig. 23 shows the detail of the
girder end. The seat length SE from JRA-2002 and N from
4) TCXDVN 375: 2006, Design of structures for earthquake
22TCN-272-05 have enough length to prevent the superstructure
from departure and unseating.
5) Specification of highway bridges, Part V seismic design,
Japan Road Association, 2002.
8. CONCLUSIONS
6) Meterials design of the existing bridge structure in Vietnam.
7) AASHTO. Load and resistance factor design (LRFD)
resistance, Vietnam Ministry of Construction, 2006.
specifications for highway bridges. Washington (DC):
The earthquake events in Vietnam are investigated and, the
seismic performance of a multi-spans highway bridge near to
American Association of State Highway and Transportation
Officials (AASHTO), 1998.
Hanoi designed by Vietnam Specification is verified by Japan
Specification. Following conclusions can be obtained.
8) AASHTO. AASHTO LRFD Bridge design specifications,
3rd Edition, Washington (DC): American Association of
1) For Level 1 earthquake motion, no serious damage is
evaluated. As to the calculation results, there are no distinct
State Highway and Transportation Officials, 2004.
9) Tongxiang An, Osamu Kiyomiya, Dynamic response
analyses and model vibration tests on seismic isolating
difference between Japanese code and Vietnam code.
2) For Level 2 earthquake motion, the bearing capacity of the
pier as to shearing force and bending moment is un-sufficient.
foundation of bridge pier, Structural Eng./Earthquake Eng.,
JSCE, Vol. 23, No. 2, pp. 195-214, 2006.
However level 2 earthquake motion that is predicted in Japan
is hardly happened according to investigation of earthquake
10) Y. Frank Chen, Assessment of the current US seismic
displacement requirements for bridges in a low-moderate
records near Hanoi.
3) When Japanese code as to Level 2 is applied to design for this
bridge, retrofit of the pier is necessary. It is required that D25 is
seismic zone, Engineering Structures, Vol. 26, pp. 1365
1379, 2004.
11) Nasim K. Shattarat, Michael D. Symans, David I. McLean,
replaced to D32 for longitudinal direction reinforcement and
D16 with 300 mm spacing is replaced to D19 with 150 mm
William F. Cofer, Evaluation of nonlinear static analysis
methods and software tools for seismic analysis of highway
for spiral reinforcement.
4) Relative displacement between the superstructure and the
bridges, Engineering Structures, Vol. 30, Issue 5, pp.
1335-1345, 2008.
substructure for Level 2 is calculated by the dynamic response
12) Kyu-Sik Park, Hyung-Jo Jung, In-Won Lee, A comparative
analysis. Seat length of the pier is about 80 cm for both design
codes. The calculated values are within the designed seat
study on aseismic performances of base isolation systems
for multi-span continuous bridge, Engineering Structures,
length. The use of stopper may decrease relative displacement
and prevent the girders from unseating when the pier is not
Vol. 24, pp. 1001-1013, 2002.
13) Yusuke Ogura, Shigeki Unjoh, Response characteristics of
–
bridge abutments subjected to collision of girder during an
collapsed.
earthquake, Structural Eng./Earthquake Eng., JSCE, Vol.
23, No. 1, pp. 135s-141s, 2006.
References
1) Rabin Tuladhar, Nhu Nguyen Hong Cuong, Fumio
Yamazaki, Seismic microzonation of Hanoi, Vietnam using
14) M. Ala Saadeghvaziri, A.R. Yazdani-Motlagh, Seismic
behavior and capacity/demand analyses of three multi-span
microtremor observations, World Conference 13th on
Earthquake Engineering, Vancouver, B.C., Canada, pp.
simply supported bridges, Engineering Structures, Vol. 30,
pp. 54-66, 2008.
25-39, 2004.
15) Do Hyung Lee, Eunsoo Choi, Goangseup Zic, Evaluation of
2) Assessment of seismic hazard at the construction site of
Nhat Tan Bridge and adjacent areas, Vietnamese Academy
earthquake deformation and performance for RC bridge
piers, Engineering Structures, Vol. 27, pp. 1451-1464, 2005.
of Science and Technology, Institute of Geophysics, 2007.
3) 22TCN 272-05, Specification for bridge design, Ministry of
Transport of Vietnam, 2005.
(Received September 18, 2008)
-549-